-
Notifications
You must be signed in to change notification settings - Fork 685
[BE] Add selected custom ops to CI #11744
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/11744
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 Cancelled JobAs of commit f318b9b with merge base 5cc5421 ( CANCELLED JOB - The following job was cancelled. Please retry:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
@kimishpatel has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
This PR needs a
|
ddfd449
to
ea8cc16
Compare
@kimishpatel has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
) | ||
|
||
@unittest.skipIf( | ||
not is_fbcode(), "in OSS error is too large 0.0004 for some reason" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Let's enable in OSS. Otherwise we will never enable in OSS.
We can finetune atol and rtol during comparison.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I am little reluctant on this because on one hand yes we can make it pass by tuning this knob, but on the other hand if there is an issue we wont catch it. My preference would be to have this test at least running internally without compromise. I dont know the reason as to why it requires low atol
self.assertTrue(torch.allclose(ref_output, op_output, atol=atol)) | ||
|
||
@unittest.skipIf( | ||
not is_fbcode(), "in OSS error is too large 0.0002 for some reason" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What does ""in OSS error is too large 0.0002 for some reason" mean?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
When I ran this test in OSS error i got was 0.0002. Which is too large for this test to pass. It does pass internally so not sure if this is blas lib issue or mac vs linux. Bug I wanted to enable the tests in oss since we have to coverage otherwise
ea8cc16
to
9000e3b
Compare
9000e3b
to
8eb31e2
Compare
@kimishpatel has imported this pull request. If you are a Meta employee, you can view this in D76775724. |
Summary: Earlier custom sdpa and kv cache werent being tested in OSS CI. This diff changes that. Tests CI ghstack-source-id: 3558645 Pull Request resolved: pytorch#11743 ### Summary [PLEASE REMOVE] See [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests) for ExecuTorch PR guidelines. [PLEASE REMOVE] If this PR closes an issue, please add a `Fixes #<issue-id>` line. [PLEASE REMOVE] If this PR introduces a fix or feature that should be the upcoming release notes, please add a "Release notes: <area>" label. For a list of available release notes labels, check out [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests). ### Test plan [PLEASE REMOVE] How did you test this PR? Please write down any manual commands you used and note down tests that you have written if applicable.
Summary: Earlier custom sdpa and kv cache werent being tested in OSS CI. This diff changes that. Tests CI ghstack-source-id: 3558645 Pull Request resolved: pytorch#11743 ### Summary [PLEASE REMOVE] See [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests) for ExecuTorch PR guidelines. [PLEASE REMOVE] If this PR closes an issue, please add a `Fixes #<issue-id>` line. [PLEASE REMOVE] If this PR introduces a fix or feature that should be the upcoming release notes, please add a "Release notes: <area>" label. For a list of available release notes labels, check out [CONTRIBUTING.md's Pull Requests](https://github.com/pytorch/executorch/blob/main/CONTRIBUTING.md#pull-requests). ### Test plan [PLEASE REMOVE] How did you test this PR? Please write down any manual commands you used and note down tests that you have written if applicable.
Summary:
Earlier custom sdpa and kv cache werent being tested in OSS CI. This diff changes that.
Tests
CI
ghstack-source-id: 35586450183e53b22971190f939bd6e362b3859a
Pull Request resolved: #11743
Summary
[PLEASE REMOVE] See CONTRIBUTING.md's Pull Requests for ExecuTorch PR guidelines.
[PLEASE REMOVE] If this PR closes an issue, please add a
Fixes #<issue-id>
line.[PLEASE REMOVE] If this PR introduces a fix or feature that should be the upcoming release notes, please add a "Release notes: " label. For a list of available release notes labels, check out CONTRIBUTING.md's Pull Requests.
Test plan
[PLEASE REMOVE] How did you test this PR? Please write down any manual commands you used and note down tests that you have written if applicable.